Assessment of postoperative outcomes in spinal epidural abscess following surgical decompression

Abstract

Background contextA spinal epidural abscess (SEA) is a serious condition that may be managed with antibiotics alone or with decompressive surgery combined with antibiotics. PurposeThe objectives of this study were to assess the clinical outcomes of SEA after surgical management and to identify the patient-level factors that are associated with outcomes following surgical decompression and removal of SEA. Study design/settingRetrospective chart review analysis. Patient sampleAn analysis of 154 consecutive patients who initially presented to a tertiary-care, academic medical center with SEA, and were subsequently treated with surgery between 2010 and 2015 was performed. Outcome measuresPostoperative predischarge American Spinal Injury Association Impairment Scale (AIS) scores, 6-month follow-up encounter AIS scores, need for revision surgery, and mortality during SEA surgery were the primary outcomes.Physiological Measures: AIS scores. MethodFisher's exact and Wilcoxon rank-sum tests were used to assess the associations between patient-level factors and surgical outcomes. Moreover, an interactive, predictive model for postoperative predischarge AIS score was developed using a proportional odds regression model. There was no funding secured for this study and there is no conflict of interest-associated biases. ResultsOne hundred fifty-four patients (mean age of 58 years) were treated using surgical decompression in addition to antibiotics. The majority of patients were Caucasian (81%) and male (61%). No intraoperative mortality was reported. A second SEA surgery was performed in 8% of patients. A comparison of the preoperative and postoperative predischarge AIS scores showed that 49% of patients maintained a score of E or improved, while 45% remained at their preoperative status and 6% worsened. Among a subset of patients (n=36; 23%) for whom a 6-month follow-up encounter occurred, 75% maintained an AIS score of E or improved, 19% remained at their preoperative status, and 6% worsened. Both the presence and longer duration of preoperative paresis was associated with an increased risk of remaining at the same AIS score or worsening at the predischarge encounter (both p< .001). A predictive model for predischarge AIS scores was developed based on several patient characteristics. ConclusionsSurgical decompression can contribute to improving or maintaining AIS scores in a high percentage of SEA patients. The presence and duration of preoperative paresis are prognostic for poorer outcomes and suggest that rapid surgical intervention before paresis develops may lead to improved postoperative outcomes. Our modeling tool enables an estimation of probabilities of patients’ predischarge condition.